The invention relates to a parallel hybrid electric vehicle energy management method based on self-supervised learning, and belongs to the technical field of hybrid electric vehicle control. According to the method, a hierarchical reinforcement learning structure is provided by using a self-supervised learning method, so that the problem of sparse rewards is solved, different working conditions can be adapted while the fuel economy of the vehicle is improved, and optimized energy control of the HEV is realized. Compared with a traditional reinforcement learning algorithm, the management method has the advantages that the fuel economy is improved; the provided algorithm is close to the optimal fuel consumption curve, and when the rotating speed is increased, it can be guaranteed that the engine works in a low-fuel-consumption and high-power area as much as possible; the SOC value of the battery can be better maintained, the charging and discharging times are reduced, and the service life of the battery is ensured.
本发明涉及一种基于自我监督学习的并联式混合动力汽车能量管理方法,属于混合动力汽车控制技术领域。本发明用自监督学习的方法,提出一种分层强化学习结构,用来解决稀疏奖励的问题,提高车辆的燃油经济性的同时,能适应不同的工况,实现HEV的优化能量控制。本发明的管理方法相较于传统强化学习算法提高了燃油经济性;提出的算法接近最佳燃油消耗曲线,在转速增大时,能尽力保证发动机在低油耗,高功率区工作;能更好的维护电池SOC值,减少充放电次数,保证了电池的使用寿命。
Parallel hybrid electric vehicle energy management method based on self-supervised learning
基于自我监督学习的并联式混合动力汽车能量管理方法
2022-06-28
Patent
Elektronische Ressource
Chinesisch
IPC: | B60W CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION , Gemeinsame Steuerung oder Regelung von Fahrzeug-Unteraggregaten verschiedenen Typs oder verschiedener Funktion |
Europäisches Patentamt | 2022
|Energy management method of hybrid electric vehicle based on power spectrum self-learning prediction
Europäisches Patentamt | 2015
|Hybrid electric vehicle energy management method based on reinforcement learning
Europäisches Patentamt | 2021
|Europäisches Patentamt | 2022
|SAE Technical Papers | 2023
|